NITROGEN CONTENT OF BACTERIAL CELLS.

I. METHOD.

BY HAROLD C. BRADLEY AND M. STARR NICHOLS.

(From the Laboratory of Physiological Chemistry and the State Laboratory

of Hygiene, University of Wisconsin, Madison.)

(Received for publication, January 5, 1918.)

The production of acids, gases, and intermediate products, of

Downloaded from http://www.jbc.org/ by guest on March 28, 2019

toxins, agglutinins, and many other protein complexes serves to
differentiate strains and groups of organisms and at the same
time furnishes us with considerable interesting data in regard to
bacterial metabolism. On the other hand, we have very little
information as to the proximate and ultimate composit,ion of
bacteria. The difficulties attending the growth of the organism
in sufficient quantity for accurate analysis and the difficulties of
the analytical technique have largely prevented the development
of this aspect of the work.
We know that higher plants differ from each other in their
nitrogen content, and in the compounds characteristic of each.
Enough work has already been done to indicate that wide differ-
ences may exist in the nitrogen content of bacteria. Nicolle and
Alilaire (1) and Wheeler (2) give the results tabulated below for
the nitrogen content of various organisms.
The results show the wide variation in the nitrogen content of
various organisms and also the great discrepancies between
analyses of the same organism by different persons. These re-
sults are of doubtful value, probably because of the variety of
conditions maintained during growth, and because the technique
allows errors of considerable magnitude where such small
amounts of material were to be used. It was decided to apply
the Folin method (3) to our problem and after certain adaptations
we arrived at the following method for determining the nitrogen
content of bacteria.
525
 Nitrogen in Cells

i-- Nitrogen.
Kind of organism.
Nicolle and
Wheeler.
AJi!aire.
-
per cant Per cent
B. proteus.vutgaris. ............................ 10.73 6.791
Typhoid ....................................... 8.28 11.55
Anthrax ....................................... 9.22 10.28
Colon .......................................... 10.32 10.655
B. pyocyaneus .................................. 9.79 10.843
‘( subtilis ...................................... - 5.964

Downloaded from http://www.jbc.org/ by guest on March 28, 2019

“ violaceus .................................... - 11.765
“ tuberculosis “air-dried”. .................... - 9.27
Chlorella vulgaris ............................... 3.96 -
B. dysenterice liquefaciem ....................... 8.89 -
Li pneumonire .................................. 8.33 -
” psittacosis ................................... 9.55 -
Glanders ....................................... 10.47 -
Sarcina aurantiaca ............................. - 11.46
-
Method.
A battery of ten hard glass test-tubes was set up together
with a sand bath and Bunsen burners to furnish the required tem-
perature. Inverted funnels with bent stems connected with
rubber tubes to a suction bottle were placed over the tubes to
remove the fumes and exclude the dust.
The bacteria used in the experiment were cultures of Bacillus
diphtherice which had been isolated from the throats of persons
suffering from the disease. These bacteria were grown on Loef-
fler’s blood serum medium slants for 72 hours. It usually re-
quired eight of these tubes to furnish enough material for one
determination. The growth of organisms was removed without
disturbing the solid medium, by a glass spade with rounded edges,
made by flattening a small glass rod. The material was spread
on a tared cover-slip and dried in a calcium chloride desiccator
at 37°C. for 72 hours.
On account of the small amount of the material obtainable
from a reasonable number of tubes, the weighing was done on
an assay balance which was sensitive to 0.000005 gm. The
weighing glasses were the thin, 15 mm. square cover-slips used
in bacteriological work. They were cleaned in concentrated
 H. C. Hradley and AI. S. Nichols 527
nitric acid, washed in distilled water several times, and then
finally in alcohol, out of which they were wiped dry when used.
Into each of the tcrl tubes was placed the following charge:
!m.
Potassium slllfatc........................................... 2.0
Copper sulfxte.............................................. 0.2
Sulfuric acid, c0ncentrat.d.. .. . ... 5.0
‘\Veighect cover-slip and materinl.

The digwtinn was continwd for 15 rninutvs after tht: liquid

had lxcorr~c colorlcm. In practically all the material digested,

Downloaded from http://www.jbc.org/ by guest on March 28, 2019

thcrc was noticed a small quantity of material which digested
with diffkultp. After the dig& had cooled somcn-hat it was
diluted with n-atcr, made alkaline with sodium hydrosidc, and
TABLE I.

Rnl. gm. c7m. par colt

0 080280 0.004665 0.000370 7.9
0 081060 0 004GiO 0.000390 8.3

0.076380 0.005730 0.00048<5 8 -5

0 OTGQ25 / O.OOiO% 0.000575 S.l

1 0.003360 0.000290 8.B

I 0.005420 0.000440 8.2

0.007300 0.0005Gl s.0

!, 0.004040 0.000357 8.8
I
0 056470 0 0028SO 0.00024;? 8.9
0 0;9020 0.00%310 0.000413 5.7
I
0.073330 I 0.001485 0.000120 8.4
0 OSO4GO 0 00257.5 0.000230 9.7*

0.097390 0.0033QO 0.000201 8.G

O.O8QO7*5 0.001350 0.000154 11.4*

0 063480 0.006170 0.000470 7.6

0.068180 I 0 004625 0.000490 10.5*

* On acconnt of the great v-:triunce from the nvcrage it seems permis-

sible to omit these results from thr average.
528 Nit.rogen in Cells
aspirated with ammonia-free air into 0.1 N sulfuric acid. The
acid solution containing the ammonium sulfate was Scsslerized
and compnrcd with a standard solution of ammonium sulfate
similarly Kesslerized, by means of the Duboscq colorimetcr. A
convcnicnt strength of standard to use, if the quantity digested
is about 5 mg., contains an ammonia equivalent of 0.0000025 gm.
of nitrogen per cc. With every set, of determinations two blanks
wcrc run and t,he average of these was subtxacted from the nitrogen
found.
The results of sixteen separate determinations are given in

Downloaded from http://www.jbc.org/ by guest on March 28, 2019

Tnblc I.
A nitrogen dctcrmination was made on the Hoffmann ba-
cillus, a diphthcroid organism, by this method with the results
as recorded in Table II.
TABLE II.

Nitrogen Content of Bacillus hoffmanni.

-
Cover-glass and dry Dry bacteria. Net weight of
Nitrogen.
bacteria. nitrogen.

gm. !7m. urn. pm cent

o.ot4110 0.002900 0.000277 9.6
0.075590 0.003730 0.000356 9.5

0.100210 0.006235 0.000600 9.6

0.071810 0.009890 0.001020 10.3
-
In order to dctcrmine the accuracy of the method, a macro-
Kjeldahl determinnt’ion was made on a larger quantity of the
dried diphtheria bacteria with the following results:
X0. 1.. ................................... 7.9 per cent nitrogen.
“ 2 ...................................... 8.1 “ “ “

SUMMARY.

In summing up the results of these determinations the following

points seem important:
1. The nitrogen content of Bacillus diphtheria: is
per cent
Regular Kjeldahl method .......................... . . . . . . 8.0
Micro-Kjeldahl method ............................ .,.... 8.38
 H. C. Bradley and M. S. Nichols 529

The nitrogen content of Bacillus hoffmanni is found to be 9.75

per cent by this method. Under like cultural conditions then,
these two organisms show marked differences in this particular.
2. The sources of error met when an analysis is made are
overcome as follows: (a) The possibility of contamination of the
material with small pieces of medium is eliminated by the use of a
molded glass spade which rubs off the growth without scratching
the medium. (b) The presence of moisture in the organism
when weighed. The organisms were dried under like conditions
which were purely arbitrary but which were found to be adequate

Downloaded from http://www.jbc.org/ by guest on March 28, 2019

for, reducing the film to constant weight. The analytical results
themselves are sufficiently alike to show that the moisture con-
tent must have been uniform for each type. (c) The error in
weighing. That this error is only apparent is seen by comparing
the accuracy of the scale used with the amount weighed. (d)
The error in ammonia obtained from reagents. This was entirely
accounted for by running measured blanks with every determi-
nation. (e) Loss of ammonia by aspiration. In the first few
practice runs there was experienced a noticeable error at this
point; however, the rate of aspiration at the start was reduced
and the total length of time prolonged which overcame this
source of error.
CONCLUSIONS.

It is possible to determine the nitrogen content of any bac-

terium which will grow on a solid medium without liquefaction
of that medium, by this method, provided as much material as 5
mg. can be obtained.
It is our intention to study further the nitrogen differences in
bacteria of different groups and of different members of the same
group.
BIBLIOGRAPHY.

1. Nicolle, M., and Alilaire, E., Note sur la production en grand des corps
bacteriens et sur leur composition chimique, Ann. Inst. Pasteur,
1909, xxiii, 547.
2. Wheeler, in Vaughan, V. C., Vaughan, V. C., Jr., and Vaughan, J. W.,
Protein Split Products in Relation to Immunity and Disease, Phila-
delphia and New York, 1913, 65.
3. Folin, O., and Farmer, C. J., A new method for the determination of
total nitrogen in urine, J. Bid. Chem., 1912, xi, 493.
NITROGEN CONTENT OF BACTERIAL
CELLS: I. METHOD
Harold C. Bradley and M. Starr Nichols
J. Biol. Chem. 1918, 33:525-529.

Access the most updated version of this article at

http://www.jbc.org/content/33/3/525.citation

Downloaded from http://www.jbc.org/ by guest on March 28, 2019

Alerts:
• When this article is cited
• When a correction for this article is posted

Click here to choose from all of JBC's e-mail

alerts

This article cites 0 references, 0 of which can be

accessed free at
http://www.jbc.org/content/33/3/525.citation.full.ht
ml#ref-list-1